The goal of this research is to demonstrate in a canine model retroviral- mediated gene transfer as therapy for purine nucleoside phosphorylase (PNP) and adenosine deaminase (ADA) deficiency. The inherited absence of either PNP or ADA is associated with severe immune deficiency, which is fatal unless treated with some form of enzyme replacement therapy. We propose to use retroviral vectors encoding the normal gene to infect autologous target cells as vehicles for gene transfer. The research employs dogs which, because of a virtual absence of red cell PNP and ADA activity, are well suited as models for these gene therapy studies. We will study three target cell types: vascular smooth muscle cells, skin fibroblasts and bone marrow stem cells, chosen for ease of access, culture and transplantability. The specific aims are: (1) Construct improved selectable retroviral vectors encoding human PNPcDNA or ADAcDNA under the control of various promoters and enhancers. (2) Infect cultured vascular smooth muscle cells. Transplant infected selected smooth muscle cells into dog carotids and veins. Use vein seeded with smooth muscle cells to construct artery-vein fistula and monitor long- term persistence and expression of transduced genes in fistula tissue. Determine localization of transplanted smooth muscle cells in vascular tissue. (3) Infect canine skin fibroblasts with vectors. Transplant infected skin fibroblasts into dogs and monitor persistence and expression of transduced gene. (4) Infect canine bone marrow cells in vitro and test for level of vector-encoded gene expression. Transplant infected bone marrow into dogs and monitor hematopoietic progenitor cells for expression of transduced genes. The overall objective of this application is to demonstrate that retroviral-mediated gene transfer can be used for long-term treatment of these immune defects and other genetic diseases.